Luka Jovanovic, I. Strumberger, N. Bačanin, M. Zivkovic, Milos Antonijevic, Peter Bisevac
Machine learning as a subset of artificial intelligence presents a promising set of algorithms for tackling increasingly complex challenges. A notable ability of this subgroup of algorithms to tackle tasks without explicit programming coupled with the expanding availability of computational resources and information transparency has made it possible to utilize algorithms to forecast prices. In recent years, cryptocurrency has increased in popularity and has seen wider adoption as a payment method. Cryptocurrency trading and mining have become a potentially very lucrative venture. However, due to the instability of cryptocurrency prices, casting accurate predictions can be quite challenging. A novel way of approaching this challenge is by tackling it through time-series forecasting. A particularly promising method for tackling this type of problem is through the utilization of long-short-term memory artificial neural networks to attain accurate prediction results. However, the forecasting accuracy of machine learning models is highly dependent on adequate hyperparameter settings. Thus, this work presents an improved variation of the arithmetic optimization algorithm, tasked with selecting the best values of a long-short term neural network casting price predictions. The presented approach has been evaluated on publicly available real-world Ethereum trading price data. The attained results of a comparative analysis against several popular metaheuristics indicate that the presented method achieved excellent results, and outperformed aforementioned algorithms in one and four-step ahead predictions.
{"title":"Tuned long short-term memory model for Ethereum price forecasting via an arithmetic optimization algorithm","authors":"Luka Jovanovic, I. Strumberger, N. Bačanin, M. Zivkovic, Milos Antonijevic, Peter Bisevac","doi":"10.3233/his-230003","DOIUrl":"https://doi.org/10.3233/his-230003","url":null,"abstract":"Machine learning as a subset of artificial intelligence presents a promising set of algorithms for tackling increasingly complex challenges. A notable ability of this subgroup of algorithms to tackle tasks without explicit programming coupled with the expanding availability of computational resources and information transparency has made it possible to utilize algorithms to forecast prices. In recent years, cryptocurrency has increased in popularity and has seen wider adoption as a payment method. Cryptocurrency trading and mining have become a potentially very lucrative venture. However, due to the instability of cryptocurrency prices, casting accurate predictions can be quite challenging. A novel way of approaching this challenge is by tackling it through time-series forecasting. A particularly promising method for tackling this type of problem is through the utilization of long-short-term memory artificial neural networks to attain accurate prediction results. However, the forecasting accuracy of machine learning models is highly dependent on adequate hyperparameter settings. Thus, this work presents an improved variation of the arithmetic optimization algorithm, tasked with selecting the best values of a long-short term neural network casting price predictions. The presented approach has been evaluated on publicly available real-world Ethereum trading price data. The attained results of a comparative analysis against several popular metaheuristics indicate that the presented method achieved excellent results, and outperformed aforementioned algorithms in one and four-step ahead predictions.","PeriodicalId":88526,"journal":{"name":"International journal of hybrid intelligent systems","volume":"19 1","pages":"27-43"},"PeriodicalIF":0.0,"publicationDate":"2023-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83757845","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The study presents an overview of theories related to e-learning barriers and e-learning behavior. Research and synthesize relevant studies at home and abroad. From related studies, identify barrier factors affecting the online learning behavior of students. Then, the research model and hypotheses for the study are presented. In this study, the author identified 5 barriers affecting students’ online learning behavior in Ho Chi Minh City: economic barriers (ECOB), interaction barriers (IB), psychological barrier (PB), environmental barriers (ENI), and regulatory institutional barriers (RIB). Previous studies revealed that using linear regression. The paper uses the optimum selection by Bayesian consideration for e-learning barriers and e-learning behavior. Get the results, then make recommendations and solutions to help educational administrators reduce barriers to increase students’ effectiveness in online learning in a better way.
{"title":"Bayesian model selection for barriers in online learning behaviors","authors":"B. Khoi","doi":"10.3233/his-230001","DOIUrl":"https://doi.org/10.3233/his-230001","url":null,"abstract":"The study presents an overview of theories related to e-learning barriers and e-learning behavior. Research and synthesize relevant studies at home and abroad. From related studies, identify barrier factors affecting the online learning behavior of students. Then, the research model and hypotheses for the study are presented. In this study, the author identified 5 barriers affecting students’ online learning behavior in Ho Chi Minh City: economic barriers (ECOB), interaction barriers (IB), psychological barrier (PB), environmental barriers (ENI), and regulatory institutional barriers (RIB). Previous studies revealed that using linear regression. The paper uses the optimum selection by Bayesian consideration for e-learning barriers and e-learning behavior. Get the results, then make recommendations and solutions to help educational administrators reduce barriers to increase students’ effectiveness in online learning in a better way.","PeriodicalId":88526,"journal":{"name":"International journal of hybrid intelligent systems","volume":"9 1","pages":"1-13"},"PeriodicalIF":0.0,"publicationDate":"2023-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77957617","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abdelhadi Ifleh, Azdine Bilal, Mounime El Kabbouri
Predicting future prices is challenging for both scholars and traders due to the high frequency and complexity of stock markets (SMs). The efficient market hypothesis (EMH) states that stock prices (SPs) follow a random walk and are unpredictably fluctuating. Furthermore, the price contains all accessible data, and we can’t extrapolate profitability from previous or current data, thus technical analysis (TA) is ineffective for projecting future prices. Technical indicators (TI) are calculated using past prices, and they are divided into two categories: trend TI and oscillators. The purpose of this study is to evaluate the accuracy of predictions for three stocks traded on the Casablanca Stock Exchange (CSE): IAM, Attijari Wafa Bank (ATW), and Banque Centrale Populaire (BCP). We combined trend TI with Long Short Term Memory model (LTSM) to make predictions and compared the results to the Random Forest model (RF). We also use Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE) to assess prediction accuracy. As a result, LSTM outperforms the RF model in terms of prediction.
由于股票市场的高频率和复杂性，预测未来的价格对学者和交易者来说都是一个挑战。有效市场假说(EMH)指出，股票价格(SPs)遵循随机漫步，并且不可预测地波动。此外，价格包含了所有可获得的数据，我们不能从以前或当前的数据推断盈利能力，因此技术分析(TA)对于预测未来的价格是无效的。技术指标(TI)是使用过去的价格来计算的，它们分为两类:趋势TI和振荡指标。本研究的目的是评估在卡萨布兰卡证券交易所(CSE)交易的三只股票的预测准确性:IAM, Attijari Wafa Bank (ATW)和Banque Centrale Populaire (BCP)。我们结合趋势TI和长短期记忆模型(LTSM)进行预测，并将结果与随机森林模型(RF)进行比较。我们还使用均方误差(MSE)、均方根误差(RMSE)和平均绝对误差(MAE)来评估预测的准确性。因此，LSTM在预测方面优于RF模型。
{"title":"Comparative study of Moroccan stock price prediction with trend technical indicators","authors":"Abdelhadi Ifleh, Azdine Bilal, Mounime El Kabbouri","doi":"10.3233/his-230002","DOIUrl":"https://doi.org/10.3233/his-230002","url":null,"abstract":"Predicting future prices is challenging for both scholars and traders due to the high frequency and complexity of stock markets (SMs). The efficient market hypothesis (EMH) states that stock prices (SPs) follow a random walk and are unpredictably fluctuating. Furthermore, the price contains all accessible data, and we can’t extrapolate profitability from previous or current data, thus technical analysis (TA) is ineffective for projecting future prices. Technical indicators (TI) are calculated using past prices, and they are divided into two categories: trend TI and oscillators. The purpose of this study is to evaluate the accuracy of predictions for three stocks traded on the Casablanca Stock Exchange (CSE): IAM, Attijari Wafa Bank (ATW), and Banque Centrale Populaire (BCP). We combined trend TI with Long Short Term Memory model (LTSM) to make predictions and compared the results to the Random Forest model (RF). We also use Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE) to assess prediction accuracy. As a result, LSTM outperforms the RF model in terms of prediction.","PeriodicalId":88526,"journal":{"name":"International journal of hybrid intelligent systems","volume":"25 1","pages":"15-26"},"PeriodicalIF":0.0,"publicationDate":"2023-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86377682","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yahia Amoura, Santiago Torres, J. Lima, Ana I. Pereira
The exponential growth in energy demand is leading to massive energy consumption from fossil resources causing a negative effects for the environment. It is essential to promote sustainable solutions based on renewable energies infrastructures such as microgrids integrated to the existing network or as stand alone solution. Moreover, the major focus of today is being able to integrate a higher percentages of renewable electricity into the energy mix. The variability of wind and solar energy requires knowing the relevant long-term patterns for developing better procedures and capabilities to facilitate integration to the network. Precise prediction is essential for an adequate use of these renewable sources. This article proposes machine learning approaches compared to an hybrid method, based on the combination of machine learning with optimisation approaches. The results show the improvement in the accuracy of the machine learning models results once the optimisation approach is used.
{"title":"Hybrid optimisation and machine learning models for wind and solar data prediction","authors":"Yahia Amoura, Santiago Torres, J. Lima, Ana I. Pereira","doi":"10.3233/his-230004","DOIUrl":"https://doi.org/10.3233/his-230004","url":null,"abstract":"The exponential growth in energy demand is leading to massive energy consumption from fossil resources causing a negative effects for the environment. It is essential to promote sustainable solutions based on renewable energies infrastructures such as microgrids integrated to the existing network or as stand alone solution. Moreover, the major focus of today is being able to integrate a higher percentages of renewable electricity into the energy mix. The variability of wind and solar energy requires knowing the relevant long-term patterns for developing better procedures and capabilities to facilitate integration to the network. Precise prediction is essential for an adequate use of these renewable sources. This article proposes machine learning approaches compared to an hybrid method, based on the combination of machine learning with optimisation approaches. The results show the improvement in the accuracy of the machine learning models results once the optimisation approach is used.","PeriodicalId":88526,"journal":{"name":"International journal of hybrid intelligent systems","volume":"11 1","pages":"45-60"},"PeriodicalIF":0.0,"publicationDate":"2023-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90937843","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.1007/978-3-031-27409-1
{"title":"Hybrid Intelligent Systems: 22nd International Conference on Hybrid Intelligent Systems (HIS 2022), December 13–15, 2022","authors":"","doi":"10.1007/978-3-031-27409-1","DOIUrl":"https://doi.org/10.1007/978-3-031-27409-1","url":null,"abstract":"","PeriodicalId":88526,"journal":{"name":"International journal of hybrid intelligent systems","volume":"30 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82772561","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Nowadays, using mobile computing devices and the Internet of Things (IoT) in networks have posed several challenges to match up the forthcoming technological requirements. Wireless Sensors Network (WSN) is considered as an important component of the IoT which produces a massive amount of data (big data). However, dealing with limited capacities of the elementary components of a network in an IoT enabled WSN, is a key challenge. The existing approaches in the literature are inadequate for large networks and cannot be applied to IoT platform without adaptation. Data Centric Network (DCN) is an important notion for the future Internet architecture to resolve the problems related to big data manipulation. In fact, using a DCN strategy for the resource limited capacities WSN enabled IoT networks is beneficial to manage densely deployed and mobile components to enhance the data gathering mechanism. In this context, this paper proposes an IoT cluster based routing protocol for data centric mobile wireless sensors networks named ICMWSN. The proposed algorithm fits with a WSN belonging a large number of mobile sensors as well as a mobile sink. It is based on a clustering technique to form multi-hops clusters around fixed pre-elected CHs. Besides, an effective tour construction method is involved for the mobile sink to collect data from the eventual cluster heads. The extensive simulation results proved that ICMWSN outperformed the compared methods in terms of energy consumption, network lifetime and data delivery rate.
{"title":"An hybrid cluster-based data centric routing protocol assisted by mobile sink for IoT system","authors":"Omnia Mezghani, Mahmoud Mezghani","doi":"10.3233/his-220012","DOIUrl":"https://doi.org/10.3233/his-220012","url":null,"abstract":"Nowadays, using mobile computing devices and the Internet of Things (IoT) in networks have posed several challenges to match up the forthcoming technological requirements. Wireless Sensors Network (WSN) is considered as an important component of the IoT which produces a massive amount of data (big data). However, dealing with limited capacities of the elementary components of a network in an IoT enabled WSN, is a key challenge. The existing approaches in the literature are inadequate for large networks and cannot be applied to IoT platform without adaptation. Data Centric Network (DCN) is an important notion for the future Internet architecture to resolve the problems related to big data manipulation. In fact, using a DCN strategy for the resource limited capacities WSN enabled IoT networks is beneficial to manage densely deployed and mobile components to enhance the data gathering mechanism. In this context, this paper proposes an IoT cluster based routing protocol for data centric mobile wireless sensors networks named ICMWSN. The proposed algorithm fits with a WSN belonging a large number of mobile sensors as well as a mobile sink. It is based on a clustering technique to form multi-hops clusters around fixed pre-elected CHs. Besides, an effective tour construction method is involved for the mobile sink to collect data from the eventual cluster heads. The extensive simulation results proved that ICMWSN outperformed the compared methods in terms of energy consumption, network lifetime and data delivery rate.","PeriodicalId":88526,"journal":{"name":"International journal of hybrid intelligent systems","volume":"146 1","pages":"137-148"},"PeriodicalIF":0.0,"publicationDate":"2022-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76573168","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
María Beatríz Bernábe Loranca, Carlos Guillén Galván, Rogelio González Velázquez, Gerardo Martínez Guzman, Alberto José Luís Carrillo Canán
The clustering of spatial-geographic units, zones or areas has been used to solve problems related to Territorial Design. Clustering adapts to the definition of territorial design for a specific problem, which demands spatial data processing under clustering schemes with topological requirements for the zones. For small sized instances, once the geographical compactness is attended to as an objective function, this problem has been solved by exact methods with an acceptable response time. However, for larger instances and due to the combinatory nature of this problem, the computational complexity increases and the use of approximated methods becomes a necessity, in such a way that when geographical compactness was the only cost function a couple of approximated methods were incorporated giving satisfactory results. A particular case of this kind of problems that has had our attention in recent years is the classification by partitioning of AGEBs (Basic Geographical Units by its initials in Spanish). Some work has been made related to the formation of compact groups of AGEBs, but additional re-strictions were often not considered. A very interesting and highly demanded application problem is to extend the compact clustering to form groups with for some of its descriptive variables. This problem translates to a multi-objective approach that has to pursue two objectives to attain a balance between them. At this point, to reach a set of non-dominated and non-comparable solutions, a method has been included that allows obtaining the Pareto front through the Hasse diagram, which implies proposing a mathematical programming model and the synthetic resulting between compactness and homogeneity.
{"title":"A bi-objective model for territorial design","authors":"María Beatríz Bernábe Loranca, Carlos Guillén Galván, Rogelio González Velázquez, Gerardo Martínez Guzman, Alberto José Luís Carrillo Canán","doi":"10.3233/his-220011","DOIUrl":"https://doi.org/10.3233/his-220011","url":null,"abstract":"The clustering of spatial-geographic units, zones or areas has been used to solve problems related to Territorial Design. Clustering adapts to the definition of territorial design for a specific problem, which demands spatial data processing under clustering schemes with topological requirements for the zones. For small sized instances, once the geographical compactness is attended to as an objective function, this problem has been solved by exact methods with an acceptable response time. However, for larger instances and due to the combinatory nature of this problem, the computational complexity increases and the use of approximated methods becomes a necessity, in such a way that when geographical compactness was the only cost function a couple of approximated methods were incorporated giving satisfactory results. A particular case of this kind of problems that has had our attention in recent years is the classification by partitioning of AGEBs (Basic Geographical Units by its initials in Spanish). Some work has been made related to the formation of compact groups of AGEBs, but additional re-strictions were often not considered. A very interesting and highly demanded application problem is to extend the compact clustering to form groups with for some of its descriptive variables. This problem translates to a multi-objective approach that has to pursue two objectives to attain a balance between them. At this point, to reach a set of non-dominated and non-comparable solutions, a method has been included that allows obtaining the Pareto front through the Hasse diagram, which implies proposing a mathematical programming model and the synthetic resulting between compactness and homogeneity.","PeriodicalId":88526,"journal":{"name":"International journal of hybrid intelligent systems","volume":"10 1","pages":"149-160"},"PeriodicalIF":0.0,"publicationDate":"2022-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89535751","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
COVID-19 is a contagious respiratory illness that can be passed from person to person. Because it affects the lungs, damages blood arteries, and causes cardiac problems, COVID-19 must be diagnosed quickly. The reverse transcriptase polymerase chain reaction (RT-PCR) is a method for detecting COVID-19, but it is time consuming and labor expensive, as well as putting the person collecting the sample in danger. As a result, clinicians prefer to use CT scan and Xray images. COVID-19 classification can be done manually, however AI makes the process go faster. AI approaches include image processing, machine learning, and deep learning. An AI model is required to diagnose COVID-19, and a dataset is necessary to train that model. A dataset consists of the information from which the model is trained. This paper consists of the review of different image processing, machine learning and deep learning papers proposed by different researchers. As well as models based on deep learning and pretrained model using gradient boosting algorithm The goal of this paper is to provide information for future researchers to work with.
{"title":"Soft computing and image processing techniques for COVID-19 prediction in lung CT scan images","authors":"Neeraj Venkatasai L. Appari, Mahendra G. Kanojia","doi":"10.3233/his-220009","DOIUrl":"https://doi.org/10.3233/his-220009","url":null,"abstract":"COVID-19 is a contagious respiratory illness that can be passed from person to person. Because it affects the lungs, damages blood arteries, and causes cardiac problems, COVID-19 must be diagnosed quickly. The reverse transcriptase polymerase chain reaction (RT-PCR) is a method for detecting COVID-19, but it is time consuming and labor expensive, as well as putting the person collecting the sample in danger. As a result, clinicians prefer to use CT scan and Xray images. COVID-19 classification can be done manually, however AI makes the process go faster. AI approaches include image processing, machine learning, and deep learning. An AI model is required to diagnose COVID-19, and a dataset is necessary to train that model. A dataset consists of the information from which the model is trained. This paper consists of the review of different image processing, machine learning and deep learning papers proposed by different researchers. As well as models based on deep learning and pretrained model using gradient boosting algorithm The goal of this paper is to provide information for future researchers to work with.","PeriodicalId":88526,"journal":{"name":"International journal of hybrid intelligent systems","volume":"47 1","pages":"111-131"},"PeriodicalIF":0.0,"publicationDate":"2022-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84135185","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The present paper aims to propose a new information-theoretic method to minimize and maximize selective information repeatedly. In particular, we try to solve the incomplete information control problem, where information cannot be fully controlled due to the existence of many contradictory factors inside. For this problem, the cost in terms of the sum of absolute connection weights is introduced for neural networks to increase and decrease information against contradictory forces in learning, such as error minimization. Thus, this method is called a “cost-forced” approach to control information. The method is contrary to the conventional regularization approach, where the cost has been used passively or negatively. The present method tries to use the cost positively, meaning that the cost can be augmented if necessary. The method was applied to an artificial and symmetric data set. In the symmetric data set, we tried to show that the symmetric property of the data set could be obtained by appropriately controlling information. In the second data set, that of residents in a nursing home, obtained by the complicated procedures of natural language processing, the experimental results confirmed that the present method could control selective information to extract non-linear relations as well as linear ones in increasing interpretation and generalization performance.
{"title":"Cost-forced and repeated selective information minimization and maximization for multi-layered neural networks","authors":"R. Kamimura","doi":"10.3233/his-220008","DOIUrl":"https://doi.org/10.3233/his-220008","url":null,"abstract":"The present paper aims to propose a new information-theoretic method to minimize and maximize selective information repeatedly. In particular, we try to solve the incomplete information control problem, where information cannot be fully controlled due to the existence of many contradictory factors inside. For this problem, the cost in terms of the sum of absolute connection weights is introduced for neural networks to increase and decrease information against contradictory forces in learning, such as error minimization. Thus, this method is called a “cost-forced” approach to control information. The method is contrary to the conventional regularization approach, where the cost has been used passively or negatively. The present method tries to use the cost positively, meaning that the cost can be augmented if necessary. The method was applied to an artificial and symmetric data set. In the symmetric data set, we tried to show that the symmetric property of the data set could be obtained by appropriately controlling information. In the second data set, that of residents in a nursing home, obtained by the complicated procedures of natural language processing, the experimental results confirmed that the present method could control selective information to extract non-linear relations as well as linear ones in increasing interpretation and generalization performance.","PeriodicalId":88526,"journal":{"name":"International journal of hybrid intelligent systems","volume":"1 1","pages":"69-95"},"PeriodicalIF":0.0,"publicationDate":"2022-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88621457","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Subrato Bharati, M. Mondal, Prajoy Podder, V. Prasath
Federated learning (FL) refers to a system in which a central aggregator coordinates the efforts of several clients to solve the issues of machine learning. This setting allows the training data to be dispersed in order to protect the privacy of each device. This paper provides an overview of federated learning systems, with a focus on healthcare. FL is reviewed in terms of its frameworks, architectures and applications. It is shown here that FL solves the preceding issues with a shared global deep learning (DL) model via a central aggregator server. Inspired by the rapid growth of FL research, this paper examines recent developments and provides a comprehensive list of unresolved issues. Several privacy methods including secure multiparty computation, homomorphic encryption, differential privacy and stochastic gradient descent are described in the context of FL. Moreover, a review is provided for different classes of FL such as horizontal and vertical FL and federated transfer learning. FL has applications in wireless communication, service recommendation, intelligent medical diagnosis system and healthcare, which we review in this paper. We also present a comprehensive review of existing FL challenges for example privacy protection, communication cost, systems heterogeneity, unreliable model upload, followed by future research directions.
{"title":"Federated learning: Applications, challenges and future directions","authors":"Subrato Bharati, M. Mondal, Prajoy Podder, V. Prasath","doi":"10.3233/HIS-220006","DOIUrl":"https://doi.org/10.3233/HIS-220006","url":null,"abstract":"Federated learning (FL) refers to a system in which a central aggregator coordinates the efforts of several clients to solve the issues of machine learning. This setting allows the training data to be dispersed in order to protect the privacy of each device. This paper provides an overview of federated learning systems, with a focus on healthcare. FL is reviewed in terms of its frameworks, architectures and applications. It is shown here that FL solves the preceding issues with a shared global deep learning (DL) model via a central aggregator server. Inspired by the rapid growth of FL research, this paper examines recent developments and provides a comprehensive list of unresolved issues. Several privacy methods including secure multiparty computation, homomorphic encryption, differential privacy and stochastic gradient descent are described in the context of FL. Moreover, a review is provided for different classes of FL such as horizontal and vertical FL and federated transfer learning. FL has applications in wireless communication, service recommendation, intelligent medical diagnosis system and healthcare, which we review in this paper. We also present a comprehensive review of existing FL challenges for example privacy protection, communication cost, systems heterogeneity, unreliable model upload, followed by future research directions.","PeriodicalId":88526,"journal":{"name":"International journal of hybrid intelligent systems","volume":"72 1","pages":"19-35"},"PeriodicalIF":0.0,"publicationDate":"2022-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74788201","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：